菜单
Analyze existing skills, agents, commands before creating new ones. Prevents duplicates and suggests alternatives.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | creation-guard |
| description | Analyze existing skills, agents, commands before creating new ones. Prevents duplicates and suggests alternatives. |
| use_when | Claude is about to create any new skill, agent, slash command, or CLI tool. MUST be invoked before writing any new artifact definition. |
Prevents duplicate functionality and enforces analysis before creating new Claude Code artifacts.
Before creating ANY new:
~/.claude/skills/*/SKILL.md)~/.claude/agents/*.md)~/.claude/commands/*.md)This skill MUST be invoked to:
Invoke this skill when you detect:
Extract from the request:
Run these searches:
# Skills - search names and descriptions
for skill in ~/.claude/skills/*/SKILL.md; do
echo "=== $(basename $(dirname $skill)) ==="
head -20 "$skill"
done
# Agents - search all agent definitions
for agent in ~/.claude/agents/*.md; do
echo "=== $(basename $agent) ==="
head -20 "$agent"
done
# Commands - search all command definitions
for cmd in ~/.claude/commands/*.md; do
echo "=== $(basename $cmd) ==="
head -10 "$cmd"
done
Also search by keywords:
grep -ril "[keyword]" ~/.claude/skills/ ~/.claude/agents/ ~/.claude/commands/
For each potentially related artifact, assess:
| Criterion | Question |
|---|---|
| Functional overlap | Does it do the same thing? (0-100%) |
| Naming confusion | Could names be confused? |
| Extension potential | Could the proposal extend this instead? |
Based on analysis, recommend ONE of:
| Recommendation | Criteria | Action |
|---|---|---|
| PROCEED | <20% overlap, genuinely new capability | Create new artifact |
| EXTEND | 50%+ overlap with single existing artifact | Modify existing instead |
| BLOCK | Would create problematic duplication | Do not create |
════════════════════════════════════════════════════════
CREATION GUARD ANALYSIS
════════════════════════════════════════════════════════
PROPOSAL:
Type: [skill|agent|command]
Name: [proposed-name]
Purpose: [one sentence]
EXISTING ARTIFACTS ANALYZED: [count]
RELATED ARTIFACTS FOUND:
1. [artifact-name] ([type])
Purpose: [what it does]
Overlap: [X]% - [explanation]
RECOMMENDATION: [PROCEED|EXTEND|BLOCK]
RATIONALE:
[2-3 sentences explaining the recommendation]
SUGGESTED ACTION:
[Specific next step based on recommendation]
════════════════════════════════════════════════════════
Proceed with creation? (y/n)
════════════════════════════════════════════════════════
Before creating ANY new artifact, ask:
User: "Create a skill for detecting AI-generated writing patterns"
Analysis finds: antislop skill already exists with 95% functional overlap.
Recommendation: BLOCK
User: "Create a command for logging work to projects"
Analysis finds: log-to-daily skill exists (70% overlap)
Recommendation: EXTEND
User: "Create a skill for managing analytics"
Analysis finds: No existing analytics skills.
Recommendation: PROCEED
Use when about to claim work is complete, fixed, or passing. Requires running verification commands and confirming output before making any success claims. Evidence before assertions, always.
Use when encountering any bug, test failure, or unexpected behavior. Requires root cause investigation before proposing fixes.
Dynamic wisdom extraction that adapts sections to content. Use when analyzing YouTube videos, podcasts, interviews, articles, or any content where you want to capture the best insights.
Detect AI-generated writing patterns (slop). Scans for banned phrases, structural tells, and scoring.
List and restore from Minervia backups. Shows available backup timestamps and lets you restore files to previous versions. Use when you need to undo an update or recover customizations.
Update Minervia to the latest version while preserving your customizations. Creates backups, shows what's new, lets you choose how to handle conflicts. Use when you want the latest skills and improvements.